<<

High Temp. Mater. Proc. 2016; 35(7): 635–641

Pei-Xian Chen*, Shao-Jun Chu and Guo-Hua Zhang A New Method to Produce Ni–Cr Used for Stainless Production

DOI 10.1515/htmp-2015-0054 overhead) [1], respectively. It is possible to make a profit- Received March 9, 2015; accepted June 23, 2015 able stainless steel production process by reducing Abstract: A new electrosilicothermic method has been and chromium addition. Currently, there are two methods proposed in the present paper to produce Ni–Cr ferroalloy, to produce Ni–Cr stainless steel: using ferronickel instead which can be used for the production of 300 series stain- of nickel metal [2] and directly reducing chromium ore less steel. Based on this new process, the Ni–Si ferroalloy fines by in a smelting reduction–kawasaki com- is first produced as the intermediate , and then the bined blown converter instead of using ferrochromium desiliconization process of Ni–Si ferroalloy melt with chro- alloy [3]. mium concentrate is carried out to generate Ni–Cr ferroal- However, there are some problems in ferronickel loy. The content in the Ni–Si ferroalloy produced in production process. Ferronickel is a ferroalloy typically the submerged arc furnace should be more than 15 mass% containing about 70–90% and 10–30% nickel. It is (for the propose of reducing dephosphorization), in order always produced by the rotary kiln-electric furnace to make sure the phosphorus content in the subsequently (RKEF) smelting process from lateritic nickel ores. But in produced Ni–Cr ferroalloy is less than 0.03 mass%. A high , submerged arc furnace (SAF) is the most widely utilization ratio of Si and a high recovery ratio of Cr can be used furnace to produce ferronickel. Generally, to meet obtained after the desiliconization reaction between Ni–Si the requirement from the traditional stainless steel ferroalloy and chromium concentrate in the electric arc production, the ferronickel alloy is expected to be furnace (EAF)–shaking ladle (SL) process. produced with a low content of silicon so that it is expected that the ores containing low silica content or – Keywords: Ni Cr ferroalloy, stainless steel, electrosili- the reductant is less used. However, lateritic nickel ore is cothermic metallurgy, dephosphorization, desiliconization always with a high silica content so that the quantity of coke has to be controlled in a low level [4, 5], which leads Introduction to the obvious increase of resistance of furnace burden. Consequently, the operation mode with a low current and high voltage for SAF has to be required in ferronickel The growth rate of stainless steel consumption is the production process. It is very different from the tradi- highest one among all materials in the world today tional operation mode for SAF, which is used to produce (ISSF, 2009). Unfortunately, its further growth will be , or ferrochromium alloy. restricted by the resource supply and production cost. Another problem is to increase the service life of furnace The consumption of nickel and chromium ferroalloys lining. Carbon-based lining was always adopted in the accounts for most of the cost of stainless steel produc- SAF when producing silicomanganese and ferromanga- tion: for example, 62% and 16% for nickel and chromium nese alloys. However, it cannot be used in the production in the production of stainless steel slabs (not including of ferronickel alloy because a large amount of carbon could be quickly dissolved into ferroalloy with low silicon *Corresponding author: Pei-Xian Chen, School of Metallurgical and content. Therefore, if lateritic nickel ore is used to pro- Ecological Engineering, University of Science and Technology duce ferronickel with high silicon content in the tradi- Beijing, Xueyuan Road 30, Haidian District, Beijing 10083, P.R. China, E-mail: [email protected] tionally SAF with carbon-based lining, the problems Shao-Jun Chu, School of Metallurgical and Ecological Engineering, mentioned above can be avoided. University of Science and Technology Beijing, Xueyuan Road 30, In addition, utilization of chromium concentrate is one Haidian District, Beijing 10083, P.R. China, of the most important problems in ferrochromium produc- E-mail: [email protected] tion process. Over 80% of the world’s known exploitable Guo-Hua Zhang, State Key Laboratory of Advanced Metallurgy, resources are located in South Africa, in which University of Science and Technology Beijing, Xueyuan Road 30, Haidian District, Beijing 10083, P.R. China, 30% is lumpy ore and 70% is fine or friable ore [6]. E-mail: [email protected] Usually, chromium concentrate is pelletized in pelletizing 636 P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy drum firstly and then charged into a sintering furnace to thermodynamics. Thereby, in the actual case, it can enhance its strength to bear the transport and impact, take place much more easily because of the high which causes the increases in the cost and the energy silicon content and low chromium content in the initial consumption for ferrochromium production. Hence, the ferroalloy melt: beneficial way for stainless steel production is to use 3½Si þ 2Cr O ð Þ þ 6CaOð Þ ¼ 4½Cr þ 3Ca SiO ð Þ chromite concentrate directly instead of ferrochromium 1% 2 3 s s 1% 2 4 l 1 alloy. ΔG ðT ¼1; 6731; 873 KÞ¼582:343 0:017T kJ mol In view of the problems involved in the traditional ð3Þ production processes of ferronickel and ferrochromium In the new process, the ferronickel alloy containing high alloys [6], a new method is introduced to produce the silicon, which is also named as nickel–silicon ferroalloy, Ni–Cr ferroalloy in this paper, which can provide new is designed as the source of silicon. Then Cr2O3 in the sources of Ni and Cr for Ni–Cr stainless steel production chromium concentrate is reduced by nickel–silicon fer- instead of ferronickel and ferrochromium alloys. roalloy to produce Ni–Cr ferroalloy. Therefore, a process for producing 300 series stain- less steel with this method can be proposed and its flow Metallurgical fundamental diagram is shown in Figure 1. The nickel–silicon ferroal- loy is produced by sintering-SAF process and then the The new method is based on the silicothermic reduction obtained molten metal will react with chromium concen- reaction, which is always called as electrosilicothermic trates in (EAF) and shaking ladle (SL) method in the industry of ferroalloys. Actually, it has to get primary stainless steel metal which can be placed been widely applied to produce ferromanganese and fer- in argon oxygen decarburization (AOD)/vacuum oxygen rochromium alloys with the low/medium carbon content decarburization (VOD) vessel for refining finally. The in the ferroalloy industry [7]. However, there is no report main aim of electrosilicothermic process presented in in the literatures about producing stainless steel with this this paper is to get the primary stainless steel which is method. The reaction principle of electrosilicothermic similar to that in the traditional stainless steel production method is always expressed as eq. (1) and it can be process. Table 1 shows the typical composition of expressed as eq. (2) when it is used to produce Ni–Cr ferroalloy:

½þSi ðÞ!MeO ½þMe ½SiO2 ð1Þ

y½ þ yðÞ þ xðÞ 3 Si Fe 2 Cr2O3 chromium concentrate 3 CaO lime ¼ y½ þ ðÞx y 4 Cr alloy 3 CaO SiO2 slag ð2Þ

If assuming that x ¼ 2, y ¼ 1 in eq. (2), the change of standard Gibbs free energy can be calculated as shown in eq. (3) by the data from literatures [8, 9]. In the temperature range of 1,673–1,873 K, the change of standard Gibbs free energy is –611.186 to –614.634 kJ · mol−1,which indicates that reaction (3) can spontaneously happen Figure 1: Flow diagram of electrosilicothermic process for 300 series in the standard state from the view point of stainless steel production.

Table 1: Chemical compositions of 300 series stainless steel in duplex and triplex processes.

Process Name C (%) P (%) S (%) Si (%) Mn (%) Cr (%) Ni (%)

Duplex Primary steel < <. <. <..  . Refined steel <. <. <. . . . . Triplex Primary steel <. <. – <..  . Refined steel <. <. – .... P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy 637

300 series stainless steel required in the traditionally extracted from lateritic nickel ore in SAF was taken from duplex and triplex stainless steel production processes. one plant in China. As shown in Table 1, except for Fe, Cr and Ni, attention should also be paid on phosphorus and silicon in stain- less steel production. It means that if phosphors and Process silicon can be controlled in the new electrosilicothermic process, it is possible that the primary 300 series stain- In the de-P experiment, the nickel–silicon ferroalloy with less steel is produced by this new process. about 0.04–0.06 mass% P and about 5–30 mass% Si was smelted in MgO crucible by using ferronickel alloy, metal- lic silicon and ferrophosphorus as the raw materials. Then alloy samples were taken by quartz tube before Experimental adding the de-P agent. After reacting for 10 min, another alloy sample was taken again. The weight ratio of de-P In order to determine whether it is possible to produce agent to metal was about 1:10. Two kinds of de-P agent – appropriate Ni Cr ferroalloy by the new method or not, were used: the one for precipitation de-P contains about some technological experiments have been done in a 50% calcium-silicon alloy, 25% lime and 25% fluorite; the pilot-scale plant, including dephosphorization (de-P) other one for interface de-P contains about 60% lime and – and desiliconization (de-Si) experiments in nickel silicon 40% fluorite. ferroalloy melt. In the de-Si experiment, the nickel–silicon ferroal- loy after de-P was first melted in the induction furnace, and then the mixture of chromium concentrate (50–60 Apparatus and materials mass%) and lime (40–50 mass%) was added in the sur- face of alloy melt. The samples of slag and metal were Experimental works were carried out in the medium fre- taken after reacting for about 25 min. According to the quency induction furnace. The information of corresponding process shown in Figure 1, de-Si process was divided raw materials is shown in Table 2. The ferronickel alloy into two steps. The first step was pre-de-Si process, in

Table 2: Raw materials used in de-P and de-Si experiments.

Experiment Raw material Chemical composition Size/weight

De-P Fe–Ni alloy Ratio of Ni to Fe is about : (Ni: . mass%), Si: . mass%, About  kg P: . mass%, C: . mass%, Cr: .–. mass% Metallic silicon Si: . mass%, P: . mass% Ferrophosphorus P: . mass%, Fe: .% Lime CaO: . mass%, P <. mass%, < mesh SiO: . mass%

Fluorite CaF: . mass%, < mesh

SiO <. mass%, P <. mass% Calcium–silicon alloy Si: . mass%, Ca: . mass%, < mesh Fe: . mass%, P: . mass%

De-Si Ni–Si ferroalloy Ni: about  mass%, About  kg Si: about  mass%, P < . mass%, C <. mass%, S <. mass%

Chromium concentrate CrO: . mass%, FeO: . mass%, CaO: . mass%, SiO: . mass%, P <. mass%, S <. mass% Lime CaO >  mass%, P < . mass%, < mesh S <. mass%, SiO: . mass% 638 P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy whichthesiliconcontentofnickel–silicon ferroalloy than 0.035 mass%. In this case, it is necessary to adopt was decreased from 25 mass% to about 10–19 mass%; a de-P process for nickel–silicon ferroalloy as shown in the second step was the final de-Si process, in which the Figure 1. When silicon content in iron is between 5 and silicon content was decreased from about 10 mass% to 30 mass%, the melt is under the condition of strong less than 1 mass%. reduction and it is possible to adopt reductive de-P method [12, 13]. The experimental results of reductive dephosphorizations of different nickel–silicon ferroalloy Results and discussion melts are shown in Table 3 and Figure 2. It should be mentioned that the Cr content of Ni–Cr ferroalloy pro- duced by electrosilicothermic method might be higher De-P experiment in Ni–Si ferroalloy melt than that of 300 series stainless steel or other Ni–Cr stainless steel. In this case, some low phosphorus pig Since an important evaluation criterion for primary stain- iron and electrolytic nickel can be added in the stainless less steel liquid is its phosphorus content, low phos- steel production process. phorus content is required for the Ni–Cr ferroalloy. Oneoftheimportantinfluencefactorsforthereduc- In the electrosilicothermic process, if the lime is used tive de-P is silicon content of nickel–silicon ferroalloy, with the phosphorus content less than 0.01 mass%, the and the de-P efficiency (defined as the ratio of phosphorus content of Ni–Cr ferroalloy will be deter- the change of phosphorus content to its initial value) mined by that of the nickel–silicon ferroalloy, with the is increased with the increase of the silicon content. consideration of the lower than 0.01 mass% phosphorus It was found that precipitation de-P efficiency was larger in the chromium concentrates [10]. Under this condition, than 30% when silicon content of nickel–silicon ferroal- nickel–silicon ferroalloy with the phosphorus content loy was higher than 15 mass%. However, for the inter- less than 0.03 mass% can be used for 300 series stainless face de-P, only when the silicon content was more steel production. than 25 mass%, the de-P efficiency can reach the same Usually, the phosphorus content of nickel–silicon value of precipitation de-P when silicon content was ferroalloy is controlled by employing raw materials 15 mass%. In addition, because the quantity of de-P with a low phosphorus content [11]. However, the case agent in no. 13 experiment was one time more than can happen that the phosphorus content is still higher that of other experiments, the de-P efficiency of this

Table 3: The experimental results of two kinds of de-P experiments.

Method No. De-P Before de-P After de-P efficiency (%) Fex Ni Si P Fe Ni Si P

Precipitation de-P . . . . . . . . . . . . . . . . .  . . . . . . . .  . . . . . . . .  . . . . . . . .  . . . . . . . .  Interface de-P . . . . . . . . . . . . . . . . . . . . . . . . . .   . . . . . . . .  a . . . . . . . .   . . . . . . . .  b . . . . . . . .   . . . . . . . . 

Note: aThe initial content of sulfur in the nickel–silicon alloy is about 0.02 mass%, and the initial content of sulfur in nickel–silicon alloy in other de-P experiments is less than 0.01 mass%. bThe weight ratio of the de-P agent to metal is 1:5. P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy 639

80% De-Si of Ni–Si Ferroalloy reacted Precipitation dephosphorization 70% Interface dephosphorization with chromium concentrate 60% After de-Si process, the silicon content should be less 50% than 1 mass%. Meanwhile, the recovery ratio of Cr and 40% utilization ratio of Si are very important during the de-Si process. There are two processes for de-Si in the ferroal- 30% loy industry: one-step process (EAF process) and two-step 20% process (EAF–SL process). One-step process means that

Dephosphorization efficiency Dephosphorization 10% the silicon content of nickel–silicon ferroalloy is reduced for one time and the silicothermic reaction just occurs in 0% the EAF. Two-step process refers to that the silicon con- 5 1015202530 tent is reduced for two times in EAF and SL, respectively. Silicon content (mass%) If the material balance of the two processes can be cal- Figure 2: Relationship between de-P efficiency and silicon content culated theoretically with the principles of mass conser- of nickel–silicon ferroalloy with different reductive de-P vation and the chemical equilibrium of de-Si reaction experiments. between Ni–Si ferroalloy and chromium concentrate, then the recovery ratio of Cr and utilization ratio of Si can be calculated and compared. experiment was much higher. It means that it is possibly The reaction occurring in the de-Si process is shown in an effective way to increase the de-P efficiency by eq. (4), with the equilibrium constant described in eq. (5): increasing the quantity of de-P agent when the silicon content of alloy melt is constant. However, the large 3Si þ 2CrðÞ2O3 ¼ 4Cr½þ 3ðÞ SiO2 ð4Þ amount of slag used in the production process is also inappropriate. Therefore, if the silicon content of a4 a3 K ¼ Cr SiO2 ð5Þ nickel–silicon ferroalloy should be designed to be a2 a3 Cr2O3 Si higher than 15 mass%, then the phosphorus in the K a i alloy can be reduced by reductive de-P process to meet where is equilibrium constant and i is the activity of . – the requirement of stainless steel production. Li et al. [4] So activity of silicon in molten Ni Cr ferroalloy can have proved that it was possible to produce the ferro- be expressed as nickel alloy containing about 18 mass% silicon content = a2 2 3 1 a ¼ Cr a ð Þ with the sintering-SAF process. Si SiO2 6 a O K Another important influence factor for reductive de- Cr2 3 P is the initial sulfur content of the nickel–silicon alloy. Therefore, the silicon content of Ni–Cr ferroalloy may be Compared with no. 10 experiment, the de-P efficiency of decided by the basicity of slag, the distribution ratio of Cr no. 11 was decreased by a half because of its high initial in metal and slag, ignoring the temperature dependence sulfur content. The reason for this is that the slag with of K. It means that the silicon content of Ni–Cr ferroalloy

CaO and CaF2 is also a good desulfurization agent for after the reaction reached chemical equilibrium can be Ni–Si ferroalloy. Therefore, the sulfur content of nickel– expressed as in eq. (7): silicon content must be controlled strictly before reduc-  ½Cr2 tive de-P or the quantity of de-P agent should be ½Si¼f ; R ð7Þ ðCr O Þ increased during the de-P process in the case of high 2 3 sulfur content. where R ¼ðCaOÞ=ðSiO2Þ. As shown in Table 3, it is an effective method to However, the chemical equilibrium experiment was decrease the phosphorus in nickel–silicon ferroalloy by not carried out in this paper. In the practical production the reductive de-P method. However, this method can process, the de-Si reaction also cannot reach real chemical also lead to the formation of Ca3P2 in slag, which is equilibrium, so that the relationship among silicon con- 2 noxious and difficult to be handled. Fortunately, an effec- tent, R and ½Cr =ðCr2O3Þ obtained from the present tech- tive method to deal with Ca3P2-bearing slag has been nological experiments is enough to guide the de-Si in the developed elsewhere [14, 15]. stainless steel production process proposed in this paper. 640 P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy

Table 4: Compositions of metal and slag in the de-Si experiments of Ni–Si ferroalloy.

Process Metal (mass%) Slag (mass%)

Fe Si Ni Cr P S CrO CaO SiO MgO AlO FeO R

Pre-de-Si . . . . . <. . . . . . . . . . . . . <. . . . . . . . . . . . . <. . . . . . . . . . . . . <. . . . . . . . . . . . . <. . . . . . . . Final de-Si . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . <. . . . . . ..

The results of de-Si experiments are shown in Table 4, and basicity is from 1.3 to 1.6, the recovery ratio of Cr is 2 the relationship among silicon content, R and ½Cr =ðCr2O3Þ controlled at the level of 85.32–93.48%; while, when it after de-Si is expressed in eqs (8) and (9). is produced by EAF–SL process, the silicon content in the When silicon content is higher than 10 mass%, the alloy must be within 15.70–16.30% after pre-de-Si to relationship is result in a recovery ratio of Cr to be 93.95–97.53%. In other words, the EAF–SL process can increase the recov- ½ 2 ½ ¼ : : Cr% : R ð Þ ery ratio of Cr although the corresponding equipment Si% 60 703 0 030 ð Þ 29 616 8 Cr2O3% investment also increases. When silicon content is less than 10 mass%, the In addition, Table 4 shows that the phosphorus and relationship is sulfur of alloy were not increased obviously after de-Si. Figure 3 shows the change of alloy composition during ½Cr%2 ½Si%¼4:967 0:003 2:587R ð9Þ the present process. The final product after electrosili- ðCr O %Þ 2 3 cothermic process contains about 30 mass% Cr, which According to the mass conservation and eqs (8) and is higher than the requirement for 300 series stainless (9), the comparisons of the EAF and EAF–SL processes steel liquid (shown in Figure 1). It can be used as the for de-Si are calculated and shown in Table 5 [16]. When master alloy of 300 series stainless steel with the addi- Ni–Cr ferroalloy is produced in EAF process and the slag tions of a certain amount of nickel plate and iron.

Table 5: Calculation of material balance for EAF and EAF–SL processes.

Raw materials for calculation

Ni–Si–Fe Fe Si Ni Cr P S .% .% .% .% .% .%

Cr concentrate CrO FeO SiO CaO MgO AlO .% .% .% .% .% .%

Lime CaO SiO AlO MgO FeO P .% .% .% .% .% .%

Process Ni–Cr ferroalloy Disposal slag Utilization ratio of Si Recovery ratio of Cr

Ni Cr Si CrO R

EAF process .% .% .% .% . .% .% .% .% .% .% . .% .% .% .% .% .% . .% .% .% .% .% .% . .% .% EAF–SL process .% .% .% .% . .% .% .% .% .% .% . .% .% P.-X. Chen et al.: A New Method to Produce Ni–Cr Ferroalloy 641

production) with a low phosphorus content. In addition, Ferronickel alloy (Ni: 8.75%, Cr: 2.51%, Si: 2.43%, P: 0.023%, S < 0.01%) to get a high recovery ratio of Cr and a high utilization

Add ferrophosphorus and metal silicon ratio of Si, the de-Si process can be optimized by using EAF–SL process. Ni–Si ferroalloy with high P content (Ni: 6.68%, Cr: 1.87%, Si: 24.79%, P: 0.059%, S < 0.01%) Funding: The authors gratefully acknowledge the finan- Precipitation de-P Add SiCa alloy, lime and fluorite cial support from National Natural Science Foundation of China (no. 51274030) for this project. Ni–Si ferroalloy after de-P (Ni: 6.59%, Cr: 1.86%, Si: 25.57%, P: 0.030%, S < 0.01%, C: 0.11%)

De-Si process Add chromium concentrate and lime References Ni–Cr ferroalloy after de-Si (Ni: 6.18%, Cr: 30.87%, Si: 0.26%, P: 0.030%, S < 0.01%, C: 0.11%) [1] E. Fritz, Steel World, 7 (2002) 33–39. [2] T. Watanabe, S. Ono, H. Arai and T. Matsumori, Int. J. Miner. – Figure 3: Pilot-scale experiment for the Ni Cr ferroalloy production Process., 19 (1987) 173–187. with electrosilicothermic method. [3] Y. Kishimoto, K. Taoka and S. Takeuchi, Kawasaki Steel Giho, 28 (1996) 213–218. [4] C.S Li, L.G. Gu, H.K. Li and H.Y. Wang, Ferroalloys, 46 (2009) 6–10 (in Chinese). However, when the silicon content of Ni–Si ferroalloy is [5] N. Voermann, T. Gerritsen, I. Candy, F. Stober and A. Matyas, about 15 mass%, which is in accordance with the require- Proceedings of the 10th International Ferroalloys Congress, – ment of precipitation de-P process, the Cr content in the SAIMM, February 1 4, 2004, Cape Town, SAIMM, Cape Town – – (2004), pp. 455 465. final Ni Cr ferroalloy is about 20 mass%. In this case, the [6] S.J. Chu, P.X. Chen and D.H. Yang, Iron Steel, 50 (2015) 1–7 composition of Ni–Cr ferroalloy is similar to the composi- (in Chinese). tion of 300 series stainless steel. Therefore, it is possible [7] H. Jalkanen and M. Gasik, Chapter 3, in Handbook of to be used as primary stainless steel and directly fed into Ferroalloys, edited by M. Gasik, Butterworth-Heinemann, AOD or VOD. Oxford (2013), pp. 32–33. [8] G.K. Sigworth and J.F. Elliott, Meter. Soc., 8 (1974) 298–310. [9] I. Barin, transited by N. L. Chen, Thermochemical Data of Pure Substances, Science Press, Beijing (2003) (in Chinese). Summary [10] N. Haque and T. Norgate, J. Clean. Prod., 39 (2013) 220–230. In the present study, a new electrosilicothermic process [11] D.H. Yang, S.J. Chu and P.X. Chen, Ferroalloys, 48 (2011) 1–4 (in Chinese). has been discussed to produce the Ni–Cr ferroalloy [12] S.S. Shibaev, P.V. Krasovskii and K.V. Grigorovitch, ISIJ Int., which is the master alloy of stainless steel, based on 45 (2005) 1243–1247. the reaction between nickel–silicon ferroalloy and [13] S. Tabuchi and N. Sano Metall. Mater. Trans. B, 15B (1984) chromium concentrate. In order to produce the Ni–Cr 351–356. ferroalloy meeting the requirement for 300 series [14] S.L. Zeng and S.J. Chu, Ferroalloys, 47 (2010) 17–21 stainless steel production, the silicon content of (in Chinese). [15] J.H. Shin and J.H. Park, Proceedings of the 13th International nickel–silicon ferroalloy should be higher than 15 mass Ferroalloys Congress, ACMI, June 9–13, 2013, Almaty, ACMI, %. Meanwhile, the phosphorus should be controlled to Almaty (2013), pp. 575–583. be less than 0.03 mass% by reductive de-P method or [16] P.X. Chen and S.J. Chu, Ferroalloys, 52 (2015) 18–24 using the raw material (for nickel–silicon ferroalloy (in Chinese).